MySQL Enterprise High Availability

A Reference Guide

A MySQL® White Paper

© 2018, Oracle Corporation and/or its affiliates

 Table of Contents

MySQL Enterprise High Availability ............................................................................................ 1

A Reference Guide ..................................................................................................................... 1
1 Introduction ....................................................................................................................... 3
2 MySQL Enterprise Edition ................................................................................................... 3
3 Oracle MySQL Cloud Service ............................................................................................... 4
4 An Introduction to MySQL Enterprise High Availability ......................................................... 5
5 Infrastructure Considerations ............................................................................................. 8
6 Software Stack ................................................................................................................... 8
7 Reference Architectures ..................................................................................................... 9
8 Developer Best Practices .................................................................................................. 12
9 Read Scale Out ................................................................................................................. 14
10 Write Scale Out ................................................................................................................ 15
11 Fencing off Failed Instances .............................................................................................. 15
12 Backups ............................................................................................................................ 15
13 Monitoring ....................................................................................................................... 16
14 Oracle Premier Support .................................................................................................... 19
15 Conclusion ....................................................................................................................... 20
16 Additional Resources ........................................................................................................ 20

© 2018, Oracle Corporation and/or its affiliates 2

1 Introduction

Whether you are building high volume websites, enterprise and

departmental applications, or advanced communications networks, your
organization needs the tools to build and manage these business-critical
database applications. This paper explores how you can confidently deploy
and operate MySQL Enterprise HA clusters, on-premises or in the Cloud, as
part of a cost-effective cross-platform solution for delivering high-performing,
highly available, disaster tolerant, reliable, and scalable applications. It
examines some of the challenges associated with building and supporting
scalable, data-driven applications and provides a detailed overview of how
MySQL Enterprise HA can address these challenges. MySQL Enterprise
Edition combines the most secure, scalable, “always on” version of the
MySQL database with online backup, monitoring, management and visual
database design and SQL development tools, all backed by Oracle Premier
Support, 24x7 global enterprise-class support services. Further, MySQL
Enterprise Edition supports your use of MySQL in conjunction with many of
the Oracle products and tools you may already be familiar with or are
currently using. MySQL Enterprise Edition is specifically designed to help
you bring secure, high performing, highly available, and scalable MySQL
applications to market faster, mitigate risk, and ensure you meet customer
and end-user Service Level Agreements (SLAs).

2 MySQL Enterprise Edition

MySQL is the world's most popular open source database. Whether you are
a fast growing web property, software vendor, a large organization or an
SMB, MySQL can cost-effectively help you deliver high performance,
scalable, and highly available database applications. If you are currently
using MySQL, you probably started with MySQL Community Edition. In
fact, in many instances MySQL enters an organization via an application
development project and makes its way into the data center when the
application is promoted for production use. Challenges arise when these
applications become vital to business revenues or key business functions.
The most common challenges around running MySQL and other open
source technologies are revealed by a simple line of questioning:

• How will you ensure you are using the most reliable, secure, scalable,
up-to-date version?
• How will you know:
o If a server or application is down?
o If there is a clustering, data replication, or consistency issue?

© 2018, Oracle Corporation and/or its affiliates 3

 o If something is affecting the performance of a server?
• Will you know:
o If MySQL is configured to scale-out as your application data and
customer base grows?
o If MySQL can survive regional outages and disasters?
o How to configure MySQL variables to ensure your applications run
at their peak performance?
• How will you:
o Identify security loopholes in MySQL servers?
o Know when there have been security alterations on a MySQL
server?
o Seamlessly add auditing compliance to your new and existing
MySQL applications?
• How will you:
o Optimize your database designs and queries before they are
migrated into your production environments or included in your
products?
o Ensure HA clusters are configured correctly for availability,
performance, and scalability?
o Monitor and tune poorly performing user/application SQL code?
• How will you integrate MySQL with your existing security standards and
infrastructure?

To help you answer these and other critical questions with confidence,
MySQL provides MySQL Enterprise Edition. MySQL Enterprise Edition is a
commercial offering comprised of the MySQL database with high
availability, security, encryption, auditing, scalability extensions, online
backup, monitoring, management, and visual database design and SQL
development tools. MySQL Enterprise Edition is backed by Oracle Premier
support for organizations delivering highly available, business critical
applications and services.

For MySQL deployments in the cloud, Oracle also provides Oracle MySQL
Cloud Service.

3 Oracle MySQL Cloud Service

Oracle MySQL Service Cloud delivers a secure, cost-effective and

enterprise-grade MySQL database service—with HA and DR provided as
basic features of the service. Built on MySQL Enterprise Edition and
powered by the Oracle Cloud, it provides the best in class management
tools to automate administrative tasks such as provisioning, scaling,
patching, backup & recovery, monitoring, and tuning. Multi-layered security
protects your data against external attacks while helping you achieve
regulatory compliance. Self-service provisioning and automated scaling

© 2018, Oracle Corporation and/or its affiliates 4

enables users to elastically scale compute resources, storage resources,
and MySQL services.

The Oracle MySQL Cloud Service video will provide you a rapid overview:
https://www.youtube.com/embed/xXUlHvAYtiE?autoplay=1

Oracle MySQL Cloud Service makes it extremely easy to rely on MySQL

either to migrate existing on-premises applications to the cloud or deliver
new ones. You can:

• Increase Business Agility: Focusing your resources on innovation,

not on infrastructure management.

• Ensure Security, Performance & Uptime: relying on the most

comprehensive MySQL cloud platform – straight from the source.

• Reduce Total Cost of Ownership: Saving on infrastructure and

database management operations costs while improving uptime.

Get our eBook “Oracle MySQL Cloud Service; Propel innovation and time-
to-market”:
https://cloud.oracle.com/opc/paas/ebooks/Oracle_MySQL_Cloud_Service.p
df

Learn more about Oracle MySQL Cloud Service and sign up for a free trial:
https://cloud.oracle.com/mysql

4 An Introduction to MySQL Enterprise High Availability

The MySQL Database is a fully integrated transaction-safe, ACID compliant

database with commit, rollback, crash-recovery and row level locking
capabilities. The MySQL Database is a cost-effective solution for:
• High-performance scalable Web/Cloud/SaaS and E-commerce
applications
• Corporate Departmental OLTP and Data Mart applications
• Operational data stores integrated with Big Data repositories and
Stream Processing systems
• Low administration, high performance, reliable embedded database
applications

MySQL has always been famous for its very flexible Master/Slave
replication. However, MySQL users have been looking for a high availability

© 2018, Oracle Corporation and/or its affiliates 5

solution built into the MySQL Server. Unlike other products, MySQL Group
Replication delivers native, built-in HA for your MySQL databases:

• Core part of the standard MySQL Server

• Leverages proven MySQL features: InnoDB, Global Transaction
IDs (GTIDs), binary logs, multi-threaded replication execution, multi-
source replication, Performance Schema
• Supported on all MySQL platforms

Figure 8: MySQL includes a native full stack, built-in HA solution

Group Replication implements both a single-primary mode with automatic

leader election and a multi-master update everywhere mode. By using a
powerful new group communication system, which provides an in-house
implementation of the popular Paxos algorithm, the group automatically
coordinates on data replication, consistency, membership, and failure
handling. This provides all of the built-in mechanisms necessary for making
your MySQL databases highly available.

Elasticity
With Group Replication, a set of servers coordinate together to form a
group. The group membership is dynamic and servers can leave, either
voluntarily or involuntarily, and join at any time. The group will automatically
reconfigure itself as needed and ensure that any joining member is
synchronized with the group. This makes it easy to quickly scale your total
database capacity up and down as needed. Router then hides all of this
complexity from your applications and their developers, as the client
connections are automatically routed to the correct server.

© 2018, Oracle Corporation and/or its affiliates 6

Failure Detection
Group Replication implements a distributed failure detector to find and
report servers that have failed or are no longer participating in the group and
the remaining members in the group coordinate to reconfigure the
membership. Router then detects these changes and automatically adjusts
the client connection routing.

Fault Tolerance
Group Replication builds on an in-house implementation of the popular
Paxos algorithm to provide distributed coordination among servers. In order
for a group to continue to function, it requires a majority of the members to
be online and for them to form an agreement on every change. This allows
your MySQL databases to safely continue to operate without manual
intervention when failures occur, without the risk of data loss or data
corruption.

Self-Healing
If a server joins the group, it will automatically bring itself up to date by
synchronizing its state with an existing member. In the event that a server
leaves the group, for instance it was taken down for maintenance, the
remaining servers will see that it has left and will reconfigure the group
automatically. When that server later rejoins the group, it will automatically
re-synchronize with the group again.

Monitoring
MySQL Enterprise Monitor has full support for HA clusters; monitoring the
configuration, health, and performance of each node and the cluster as a
whole, providing best practice advice and alerts, along with easy to
understand visual tools that allow you to easily monitor and manage all of
your HA clusters.

HA for MySQL Client Applications using MySQL Router

MySQL Router allows you to easily migrate your standalone MySQL
instances to natively distributed and highly available clusters without
affecting existing applications. Router provides transparent client connection
routing, load balancing, and failover into your HA clusters.

Easy and Accessible using MySQL Shell

MySQL Shell provides a single intuitive, flexible, and powerful interface for
all MySQL related tasks. MySQL Shell makes it easy to create, monitor, and
manage your HA clusters -- including MySQL Router instances -- in a self-
describing natural language interface, without the need to understand low
level concepts, configuration options, or other complex aspects. MySQL
Shell makes HA clusters easy and accessible for everyone.

© 2018, Oracle Corporation and/or its affiliates 7

5 Infrastructure Considerations

§ Hardware redundancy with 3-5 physical machines

• Different motherboards
• Different memory sub-systems
• Different storage devices
§ Reliable and fast network
• Eliminate hubs, switches, and network hops
• 10G+ Ethernet end-to-end between nodes
• Dedicated network
o Avoiding noisy neighbor issues
o Avoiding general congestion issues
o Providing more predictable performance
§ Site redundancy
• 1-3 data centers (sites/AZs/ADs)
• Redundant cooling systems
• Redundant internet backbone connections
• Redundant power utility providers
§ Regional redundancy (optional)
• 1+ regional sites for disaster tolerance

6 Software Stack

• Operating System: Windows or Linux (see: MySQL 8 Platforms)

• MySQL: MySQL 8 (Group Replication, Router, Shell)

© 2018, Oracle Corporation and/or its affiliates 8

 7 Reference Architectures

Figure 9: Shared Nothing Cluster – Single Data Center

This is the most basic production ready HA setup. This provides basic
availability within a single data center and protects against any failure confined
to a single machine. In order for this to be effective, ensure that the machines
are not sharing physical resources. The OS images should be running on
different physical boxes, those boxes should be in different racks, and the
boxes would ideally have redundant systems for power, cooling, and
networking.

© 2018, Oracle Corporation and/or its affiliates 9

 Figure 10: Shared Nothing Cluster – Cross Data Center

This deployment provides additional availability. Now even if we lose Data

Center 1 (DC1) entirely, we can continue on with the cluster partition in Data
Center 2 (DC2).
• Note: If DC1 fails, a manual step is required to unblock the minority
partition in DC2
• Note: The network connection between DC1 and DC2 should be
reliable and have low latency in order to avoid node evictions.

© 2018, Oracle Corporation and/or its affiliates 10

 Figure 11: Geographically Redundant Cluster

This deployment provides disaster tolerance as the data centers are a great
distance from each other on the earth. So your database service can survive
major regional disasters without interruption and without user intervention.
• Note: because these datacenters are far apart and communicate over
a high latency connection, you would use asynchronous master/slave
replication between each site and cluster
• Note: because asynchronous replication is in use, there is some
potential for losing your most recent writes when losing a regional data
center.

© 2018, Oracle Corporation and/or its affiliates 11

 Figure 12: Active/Active Regional Multi-Data Center Setup

This setup provides high availability, disaster tolerance, and low latency
access to users around the globe.
• Note: because these datacenters are far apart and communicate over
a high latency connection, you would use asynchronous master/slave
replication between each site and cluster
• Note: because master/slave connections are used between clusters—
which have no built-in support for automatic conflict prevention,
detection, and handling—you will need to ensure that the writes are
conflict free at the application layer.

8 Developer Best Practices

• Catch all client connection errors, and re-connect. This is because

Router only does any routing today for new connections. So to be safe
you can catch all client errors and re-connect whenever you get one. In
particular, be prepared to encounter and handle the following errors:
o Error: 2006 (CR_SERVER_GONE_ERROR) -- “MySQL server
has gone away”
§ Description: this means that the backend connection was
lost. Most likely the result of a failure.
§ Action: re-connect so that Router will automatically
connect you to the current PRIMARY
o Error: 2013 (CR_SERVER_LOST) -- “Lost connection to
MySQL server during query”

© 2018, Oracle Corporation and/or its affiliates 12

 § Description: this means that the backend connection was
lost. Most likely the result of a failure.
§ Action: re-connect so that Router will automatically
connect you to the current PRIMARY
o Error: 2055 (CR_SERVER_LOST_EXTENDED) -- “Lost
connection to MySQL server at '%s', system error: %d”
§ Description: this means that the backend connection was
lost. Most likely the result of a failure.
§ Action: re-connect so that Router will automatically
connect you to the current PRIMARY
o Error: 1836 SQLSTATE: HY000 (ER_READ_ONLY_MODE) --
“Running in read-only mode”
§ Description: this means that we’re not currently connected
to a valid PRIMARY instance. You could see this if you
have an open connection to a node that was the
PRIMARY but has recently been ejected from the group
due to an ephemeral issue (e.g. temporary network
hiccup).
§ Action: re-connect so that Router will automatically
connect you to the current PRIMARY
o Error: 3101 SQLSTATE: 40000
(ER_TRANSACTION_ROLLBACK_DURING_COMMIT) –
“Plugin instructed the server to rollback the current transaction.”
§ Description: this means that a transaction failed the group
certification process, due to one or more members
detecting a potential conflict, and was thus rolled back.
§ Action: re-try your transaction
o Error: 3635 SQLSTATE: HY000
(ER_APPLIER_LOG_EVENT_VALIDATION_ERROR) – “The
table in transaction %s does not comply with the requirements
by an external plugin.”
§ This will happen if you attempt to execute DML against a
table which fails to meet the requirements—most
commonly the table is using a storage engine other than
InnoDB or it lacks a defined Primary Key.
§ Action: examine the table noted in the error log
• If you're using any sort of client connection pool, then use a custom
pool "healthcheck" query and have that run every second on each
connection in the pool and/or executed on a connection before handing
it out from the pool. Most connection pools (e.g. JDBC app server

© 2018, Oracle Corporation and/or its affiliates 13

 connection pools) provide these options, and if the healthcheck query
fails then one or all of the connections in the pool are closed. It's
common to use ‘SELECT 1’--which just ensures that you have a
database listening and responding on the other end--but in our case we
want to know that the database on the other end is also currently a
PRIMARY ONLINE member of the HA cluster. Two (5.7 compatible)
examples of specific healthcheck queries that can be used are:
-- Is the node an ONLINE R/W member of a single or multi-primary
-- cluster with a write quorum
SELECT IF( CONCAT(
(SELECT
IF( MEMBER_STATE='ONLINE' AND ((SELECT COUNT(*) FROM
performance_schema.replication_group_members WHERE MEMBER_STATE != 'ONLINE') >=
((SELECT COUNT(*) FROM performance_schema.replication_group_members)/2) = 0),
'YES', 'NO' ) FROM performance_schema.replication_group_members JOIN
performance_schema.replication_group_member_stats USING(member_id)
WHERE member_id=(SELECT @@server_uuid)
),
IF( (SELECT (SELECT GROUP_CONCAT(variable_value) FROM
performance_schema.global_variables WHERE variable_name IN ('read_only',
'super_read_only')) != 'OFF,OFF'), 'YES', 'NO'
)
) = 'YESNO', 1, 0) as endpoint_status;

-- Is the node an ONLINE PRIMARY member of a cluster
SELECT IF(
MEMBER_STATE='ONLINE' AND
(member_id=(select variable_value from performance_schema.global_status
WHERE variable_name="group_replication_primary_member")), 1, 0
) as endpoint_status
FROM performance_schema.replication_group_members INNER JOIN
performance_schema.replication_group_member_stats USING(member_id)
WHERE member_id=(SELECT @@server_uuid);

9 Read Scale Out

MySQL HA clusters are not only about availability, they can also be
leveraged for read scale out. Router makes this easy for application
developers—for your R/O workloads you connect to the R/O port and it
transparently load balances these connections across the SECONDARY
nodes in the cluster.

• Note: in a default configuration, the system is write optimized so as to

reduce the write latency on the PRIMARY node. This means, however,
that there’s a chance that reads on SECONDARY nodes could be stale
due to a (generally small) lag in the application of transactions.
• Note: you can choose to make the system read optimized—using the
group_replication_synchronous_writes option—and you pay a
penalty in write latency and throughput on the PRIMARY, but in return
you have read-your-writes guarantees across the cluster. So your
© 2018, Oracle Corporation and/or its affiliates 14
 application developers need not be concerned about any potential for
stale reads.

10 Write Scale Out

MySQL HA clusters are primarily about ensuring that your database

applications are always available. As noted, they can also be used for read
scale out. However, for write scale out more is needed. Partitioning your
data across a set of HA clusters—where each cluster manages a subset of
the data—is needed. Then you can scale your writes as needed by further
partitioning your dataset. Today this type of setup is often called a sharded
cluster. While MySQL Enterprise HA does not have native support for
sharded clusters today, it is on our roadmap. So stay tuned!

11 Fencing off Failed Instances

This is one area where you need to choose between availability and
consistency using the group_replication_exit_state_action option.
Group Replication provides two options for how to handle failed nodes:
1. Make the MySQL instance read-only
2. Kill the MySQL instance (which in some cases will allow the node to
re-join the cluster automatically)

Option 1 will allow the cluster to continue serving reads in spite of multiple
cluster failures, such as a total loss of write quorum, but it will also allow a
failed node to provide stale reads. By default, Group Replication will kill an
instance that has failed (entered the ERROR state) and exited the cluster in
order to ensure read consistency.

12 Backups

MySQL Enterprise Backup performs online "Hot", non-blocking backups of

MySQL HA clusters. Full backups can be performed while the cluster and all
its MySQL instances are online, without interrupting queries or updates.
Incremental and partial backups are supported, along with consistent Point-
in-Time Recovery (PITR), compression, and direct cloud native storage.

• Note: Backups of an HA cluster can be performed via a PRIMARY or

SECONDARY member of the cluster. But since SECONDARY
members are read-only, for backup jobs executed on a
SECONDARY no information about that backup job is recorded in the
© 2018, Oracle Corporation and/or its affiliates 15
 local backup status and history tables and thus that backup job will
not be visible to MySQL Enterprise Monitor.

13 Monitoring

MySQL developers, DBAs, and SysAdmins often find themselves having to

manage dynamic, high growth applications that require continuous uptime.
The MySQL Enterprise Monitor helps them automate the management of
their MySQL infrastructure and improve the performance and availability of
their applications.

The MySQL Enterprise Monitor is a web-based application that can manage

MySQL HA clusters within the safety of a corporate firewall or remotely in a
public cloud. MySQL Enterprise Monitor provides:

• Performance & Availability Monitoring - Continuously monitor

MySQL health and performance related metrics
• Visual Query Analysis – Monitor query performance and pinpoint
SQL code that could cause a slow-down
• InnoDB Monitoring – Monitor key InnoDB metrics that impact
MySQL performance
• Replication Monitoring – Gain visibility into the performance, and
health of all MySQL Masters and Slaves
• HA Cluster Monitoring – Monitor and manage your distributed and
highly available MySQL setups with ease
• Backup Monitoring – Ensure your online, hot backups are running
as expected
• Firewall Monitoring – Monitor MySQL Enterprise Firewall activity
and protect your MySQL servers against database attacks
• Audit Monitoring – Enforce MySQL Enterprise Audit usage across
all of your MySQL servers
• Disk Monitoring – Forecast future capacity requirements using trend
analysis and projections
• Security Monitoring – Identify and resolve security vulnerabilities
across all MySQL servers
• Operating System Monitoring – Monitor operating system level
performance metrics such as load average, CPU usage, RAM usage
and swap usage
• Access Control Lists (ACLs) – Role and group based access
controls provide an easy way for MySQL DBAs to define fine-grained
security policies for their MySQL servers

© 2018, Oracle Corporation and/or its affiliates 16

One Enterprise Dashboard for Monitoring all MySQL Assets

The Enterprise Dashboard provides real-time visibility into the performance

and availability of all your MySQL installations. You can monitor MySQL
related metrics for a single server or groups of servers, and manage all
forms of replication and clustered topologies. The Enterprise Dashboard is
designed to give you an overall view of the health of your MySQL
installations and point you to potentially critical issues that may impact the
performance and availability of your applications. A rich set of real time and
historical graphs allow you to drill down into relevant statistics and details.

Figure 10: MySQL Enterprise Overview Dashboard

Monitoring of Replicated and Clustered Topologies

The Replication and Topology Dashboards make it easier to scale out using
MySQL replication and clustering by providing industry-leading auto
detection, grouping, documenting and monitoring of all combinations of
MySQL Master/Slave Replication and MySQL HA clusters. Changes and
additions to existing topologies are also auto detected and maintained
providing you instant visibility into changes. This helps reduce the learning
curve for anyone new to MySQL Replication and MySQL HA clusters or to
scale-out and high availability environments in general.

© 2018, Oracle Corporation and/or its affiliates 17

 Figure 11: MySQL Enterprise Replication Topology view

Whether you use a single-source tree hierarchy, circular replication, HA

clusters, or a complex, multi-level, multi-source hierarchy that mixes all of
the above, the Topology view shows how your MySQL Topology is currently
operating.

Figure 12: Multi-source Replication topology view

© 2018, Oracle Corporation and/or its affiliates 18

The Topology and Replication Dashboards show context-sensitive
information for each node and node type, as well as HA. High-level identity
and throughput metrics are always displayed. For each Replication Source,
you'll see Binary Logging and GTID status, or, if enabled, statistics on Semi-
Sync Replication or Group Replication performance. For each Replica, you'll
see the status of each Replication Channel, statistics on Relay Logging,
Multi-Threaded Replication and other source configuration details.

Figure 13: MySQL Replication Dashboard Status View

14 Oracle Premier Support

Oracle offers 24x7, global support for MySQL. The MySQL Support team is
composed of seasoned MySQL developers, who are database experts and
understand the issues and challenges you face. With Oracle Premier
Support, you can lower the total cost and risk of owning your MySQL
databases, improve the return from your IT investment, and optimize the
business value of your IT solutions. MySQL support is included in the
subscription for end users, and available separately from commercial
licenses for ISVs and OEMs. Oracle Premier Support for MySQL includes
the following features:

• 24 X 7 production support
• Unlimited support incidents
• Knowledge Base
• Maintenance releases, bug fixes, patches and updates
• MySQL Consultative support
• Staffed by the most experienced MySQL Engineers in the industry
• The ability to get MySQL support in 29 languages

MySQL Consultative Support service is included in Premier Support.

MySQL Consultative Support is a proactive approach that is designed to
© 2018, Oracle Corporation and/or its affiliates 19
 help you avoid critical outages. MySQL Support Engineers advise you on
how to properly setup and tune your MySQL servers, schema, queries, and
HA clusters to maximize performance and availability. Also, by taking the
initiative to properly design and tune your MySQL database applications you
can avoid having to purchase expensive hardware for your IT infrastructure.

Learn more about Oracle Premier Support:

http://mysql.com/support/

15 Conclusion

In this paper we explored the components that make up MySQL Enterprise

High Availability. These components are designed to help you mitigate risk
and meet Service Level Agreements (SLAs) as you implement highly
available and disaster tolerant applications on MySQL. In addition to High
Availability, MySQL Enterprise Edition extends the MySQL Database to
include advanced Auditing, Security, Encryption, and Performance/Scale
features. The MySQL Enterprise Backup performs online "Hot", non-
blocking backups of your MySQL HA clusters. The Enterprise Monitor, with
its smart Advisors and Query Analyzer, proactively notify you of potential
problems and service improvement opportunities before issues impact your
applications and customers.

You can also consume MySQL Enterprise as a service in Oracle Cloud.

Built on the proven MySQL Enterprise Edition and powered by the Oracle
Cloud, Oracle MySQL Cloud Service provides a simple, automated,
integrated and enterprise-ready cloud service that enables organizations to
deliver MySQL-based applications globally at scale.

16 Additional Resources
http://www.mysql.com/

© 2018, Oracle Corporation and/or its affiliates 20

https://cloud.oracle.com/mysql

MySQL Enterprise Edition Demo:

https://www.youtube.com/watch?v=ypQh9H9Rf9w

To contact an Oracle MySQL Representative:

http://www.mysql.com/about/contact/

© 2018, Oracle Corporation and/or its affiliates 21